Improving the referral process from primary care to an AMHT

There is established evidence that discrepancy rates in referral thresholds from GPs to secondary care services as high as 10-fold exist between individual GPs. Andrew Molodynski et al looked systematically at the amount and quality of referrals to the secondary care team in order to identify any issues and develop support for referrers to make the process as smooth as possible.

Approximately one in four people in the UK will experience a mental health problem in any given year,1 and it is widely acknowledged that mental health services are overstretched and insufficiently funded. This is despite the Department of Health (DoH) estimating that mental health problems cost the UK economy between £70 billion and £100 billion per year.2 Depression alone is ranked second in the World Health Organization (WHO) global burden of disease chart, and will be number one in a decade.3,4 Reasons for this burden include sickness absence, benefit provision and lost productivity. Given the prevalence of mental health problems, their impact on society, and the current financial pressures on health and social care, appropriate resource allocation is a pressing concern. Most mental health conditions are managed in primary care, and resource allocation between primary and secondary care is important to ensure that care is given in the most appropriate setting and that there are effective pathways to specialist care for those who need it.

In 2010, the King’s Fund published a series of reports regarding the referral process from primary to secondary care.5,6 General practitioners (GPs) are generally seen as the ‘gate keepers’ to secondary care services, and the referral process has direct consequences for patient experience of care. It is also an important cost driver in the health system. The Didcot adult mental health team (AMHT) is a generic team offering assessment and treatment to adults of working age across a mixed catchment area of approximately 50 000 with pockets of significant deprivation. The majority of referrals come directly from local GPs who must decide the urgency with which patients should be seen – emergency (within four hours), urgent (within seven days), or routine (within four weeks). A qualified duty worker discusses all urgent and emergency referrals with the referrer to assess risk and agree the timeframe for review.

There is established evidence that discrepancies in referral thresholds as high as 10-fold do exist between individual GPs.6–8 Many possible reasons for this have been suggested, such as experience and culture of practice.9,10 The AMHT does not have strict referral guidelines and communication with primary care is actively encouraged. This review was undertaken to clarify how the pathway was working.

A recent King’s Fund report entitled ‘The quality of GP diagnosis and referral’ included a series of points present in a high quality referral:5
• Necessity: are patients referred as and when necessary?
• Timeliness: is this done without avoidable delay?
• Destination: are patients referred to the most appropriate destination first time?
• Process: is the process of referral a high quality one:
– Do referrals contain the necessary information in an accessible format?
– Do the GP, patient and specialist have a shared understanding of the referral?
– Is pre-referral management adequate?

We looked systematically at the amount and quality of referrals to the team in order to identify any issues and develop support for referrers to make the process as smooth as possible, to save them time and reduce delays and disruptions, which we know are frustrating for patients.


We assessed 100 consecutive referrals to the AMHT between October 2015 and January 2016. These referrals were assessed by two raters (AD and MAN). The dataset was agreed upon and set out prior to data collection. In cases where there was uncertainty, the data was discussed with AM. Ten per cent of referrals were randomly selected and entered by both raters to ensure reliability.

The following information was collected for each referral:
• Referrer name
• Urgency of referral ascribed by referrer
• Agreed urgency after discussion
• Referrals that were not accepted for assessment
• The specification of risk
• The risk determined on assessment
• Diagnosis
• The number of face-to-face follow-up appointments with the referred patient was documented. The number of zero appointments included did not attends or referral deemed inappropriate. One = initial assessment only. The further categories included two to four appointments and five or more appointments
• A series of brief recommendations for different conditions that GPs could use to assess appropriateness for referral were used with each referral letter.

The information above was entered into an excel database and analysed using descriptive statistics.

The multidisciplinary team, including the administrative team, were consulted at all stages of this project and influenced the discussion and final recommendations.


Out of the 100 referrals we analysed, 65% were labeled as routine by the referrer and 35% as urgent or emergency (Figure 1).

The majority of referrals were received from GP practices, with 33% coming from other sources, as illustrated in Figure 2. Other sources include: child and adolescent mental health services (CAMHS), complex needs service (CNS), emergency department psychiatric service (EDPS), private sector, psychology + psychological medicine services at the general hospital, improving access to psychological therapies (IAPT) services, urgent care services, self-referral, and street triage.

Figure 3 shows the urgency of referrals specified by each referring body, with 1–7 representing individual GP practices and the other section including all other referring parties. There are no significant differences across referrers.

Figure 4 shows the proportions of cases from each level of urgency receiving follow-up. It can be seen that those referred routinely are more likely to receive greater amounts of follow up.

Figure 5 shows the percentage of referrals that clearly specified risk: only 21% overall compared with 93% in the assessment letters sent to referrers.

Fourty three per cent of those referred urgently received no diagnosis of a mental health condition or remained ‘under review’. Nearly 70% of urgent referrals were in fact seen routinely following discussion with the team, while 29% of routine referrals were ‘upgraded’ to urgent by the duty worker at the AMHT, as shown in Figure 6.

Discussion and results

The data above provide a detailed review of referral behaviours to a fairly standard mental health team. Approximately two thirds of referrals were routine. The majority of referrals came from general practice. The ratio of urgent/emergency to routine referrals from all referring parties was relatively consistent (Figure 3). The 35% of patients referred urgently place significant demand on the AMHT both financially and administratively. The fact that many were downgraded suggests both that discussion ‘works’ and that guidance might help referrers to decide on urgency more accurately. Those referred urgently were less likely to have a multidisciplinary assessment with two workers (which routine referrals do), restricting the ability to make pharmacological or other decisions. One might expect that patients referred urgently would be more likely to be more unwell and/or at risk and thus remain under the care of the team for longer. The opposite was the case in this sample, with lower levels of follow-up and many not receiving any formal diagnosis. High numbers of people without mental disorders being referred urgently might raise false expectations for them and impact on the ability of the team to function efficiently. In our experience, many of those referred urgently are ‘in crisis’ and have difficulties managing strong emotions and this is a key factor in lower rates of follow up. We have regular meetings with primary care and referrers to look at these and other issues and are currently enhancing these arrangements.

Decisions regarding resource allocation and urgency of action depend primarily on risk.11 In the time period considered here we found that a low proportion of referrals clearly described ‘risk’ – only 21%. Previous studies highlight that missing information can have effects on risk assessment, patient triage, and resource allocation.12,13 In our data set, many referrals were re-graded during the triage process. This regrading, with its disruption to the pathway, would almost certainly be reduced by the provision of clearer information. Crucially it should also help prevent the high-risk situation of patients being incorrectly processed as routine when there is in fact significant risk.14,15 Including risk in referral letters should allow all parties, including the patient, to be clearer as to the purpose and urgency of the referral to the AMHT.

Greater understanding of the referral process is very important in the patient’s overall healthcare experience,16 with declined referrals or altered timescales a frequent cause of patient anxiety and uncertainty. There are already quarterly meetings with each GP practice to allow for case discussions and communication regarding positive and negative experiences surrounding the referral process from both sides. The King’s Fund state: ‘good relationships may also make it easier for GPs to seek informal advice, reducing the need for making formal referrals and avoiding duplication of tests’.5 The overarching purpose of these regular meetings is to build such relationships and improve communication.10

We concluded from the above that the referral process could be improved by some brief suggested referral criteria for common conditions, presented as a ‘how to’ type guide designed to help referrers in an increasingly complex and fragmented system. We devised a simple and visually appealing poster including these criteria that all referrers could have and see easily. Recognising that there is a subset of patients who do not meet these criteria but could legitimately be helped by alternative services, we devised a second reference poster detailing other services available and how they can be contacted directly.

Mental health problems are highly prevalent. Effective shared care between primary care, specialist services and other agencies are crucial to ensure patients’ needs are met efficiently, safely, and humanely.17 Our project identified several possible areas of improvement, and we have decided to provide concise and clear information to all referrers regarding our service and others that may be of assistance. We will enhance our practice of regular meetings in primary care to better support our colleagues and allow for even more discussion than at present. Following this, we will seek feedback from our referrers as to how they believe the system is working, and whether they have any suggestions that may enhance our common aim – a straightforward and safe pathway to allow those who need it to access specialist assessment and care.

Dr Dehghan is FY2 Doctor, Medical Sciences Division, University of Oxford, Dr Nowar is Core Trainee, Oxford Health Foundation Trust and Dr Molodynski is Consultant Psychiatrist, Oxford Health NHS Foundation Trust and Honorary Senior Clinical lecturer Oxford University.


The authors would like to thank all members of the Didcot team and especially the admin team with a special thanks to Jan Gesner in this regard.

Declaration of interests

No conflicts were declared.


1. McManus S, Meltzer H, Brugha T, et al (eds). Adult psychiatric morbidity in England, 2007: Results of a household survey. Leeds: NHS information centre for health and social care, 2009.
2. Davies SC. Chief Medical Officer’s summary. In: N Metha, ed. Annual report of the Chief Medical Officer 2013, public mental health priorities: investing in the evidence. London: Department of Health, 2013;11–19.
3. Lepine JP, Briley M. The increasing burden of depression. Neuropsychiatr Dis Treat 2011; 7(Suppl 1):3–7.
4. Ferrari AJ, Charlson FJ, Norman RE, et al. Burden of depressive disorders by country, sex, age, and year: findings from the global burden of disease study 2010. PLoS Med 2013;10(11):e1001547.
5. Foot C, Naylor C, Imison C. The quality of GP diagnosis and referral. London: The King’s Fund, 2010.
6. Imison C, et al. Referral management: lessons for success. London: The King’s Fund, 2010.
7. Ashworth M, Clement S, Sandhu J, et al. Psychiatric referral rates and the influence of on-site mental health workers in general practice. Br J Gen Pract 2002;52(474):39–41.
8. Creed F, Gowrisunkur J, Russell E, et al. General practitioner referral rates to district psychiatry and psychology services. Br J Gen Pract 1990; 40(340):450–4.
9. Chew-Graham C, Slade M, Montâna C, et al. A qualitative study of referral to community mental health teams in the UK: exploring the rhetoric and the reality. BMC Health Serv Res 2007;7:117.
10. Chew-Graham C, Slade M, Montâna C, et al. Loss of doctor-to-doctor communication: lessons from the reconfiguration of mental health services in England. J Health Serv Res Policy 2008;13:6–12.
11. Department of Health (DoH). Best practice in managing risk. Principles and evidence for best practice in the assessment and management of risk to self and others in mental health services. DoH: London, June 2007 (updated March 2009).
12. Patel NN, D’Souza J, Rocker M, et al. Prioritisation of vascular outpatient appointments cannot be based on referral letters alone. Surgeon 2008;6(3):140–3.
13. Bodek S, Ghori K, Edelstein M, et al. Contemporary referral of patients from community care to cardiology lack diagnostic and clinical detail. Int J Clin Pract 2006;60(5):595–601.
14. Hilton C, Bajaj P, Hagger M, et al. What should prompt an urgent referral to a community mental health team? Ment Health Fam Med 2008;5(4):197–201.
15. Cubbin S, Llewellyn-Jones S, Donnelly P. How urgent is urgent? Analysing urgent out-patient referrals to an adult psychiatric service. Int J Psychiatry Clin Pract 2000;4(3):233–5.
16. Grace JF, Armstrong D. Reasons for referral to hospital: extent of agreement between the perceptions of patients, general practitioners and consultants. Fam Pract 1986;3(3):143–7.
17. Singh P. Running an effective community mental health team. Adv Psychiatric Treat 2000;6:414–22. ( accessed June 2017)

Improving the referral process from primary care to an AMHT

Add yours ↓
Web design and marketing agency Leamington Spa